| 题名 | 城市化对河流生态系统的影响研究--以北京市为例 |
| 作者 | 陈向 |
| 学位类别 | 博士 |
| 答辩日期 | 2016-06 |
| 授予单位 | 中国科学院研究生院 |
| 授予地点 | 北京 |
| 导师 | 周伟奇 |
| 关键词 | 水质 Water quality 底栖硅藻 diatom assamblages 土地利用/覆盖 land use and land cover 多元分析 multiple analysis 北京 Beijing |
| 其他题名 | The influence of urbanization on urban stream ecosystem in Beijing, China |
| 学位专业 | 生态学 |
| 中文摘要 | 城市化的快速发展对城市河流水文、水质及水生生物等产生了重大影响。本论文以快速城市化的北京为研究区域，利用水化历史数据和遥感数据，研究了2000年、2005年和 2010年，北京城市、郊区和山区不同区域河流水质的时空变化特征及其与周边土地利用/覆盖类型变化的相关关系。同时，通过采样调查，定量分析了 2014年 5月-11月五环城区内城市上游河道、城市下游河道及山区对照溪流河流水质、底栖硅藻的空间特征，及其在春季、夏季和秋季的季节变化特征，进一步分析了底栖硅藻的时空分布特征与水质的响应关系。主要研究结果和结论如下： （1）不同流域及城市五环内的河流水质空间差异明显，但时间差异较小。①不同流域中，山区河流水质显著优于城市河流和郊区河流（p<0.05），郊区河流水质最差。② 2000年－2010年，北京市的水质情况整体上逐渐趋好，表现为城市河流和郊区河流的部分水质理化指标的浓度逐年下降，而山区河流部分水质理化指标的浓度有逐年上升的趋势，但不显著。③城市五环内河流水质空间差异较大，城市下游河道水质理化指标浓度显著高于城市上游河道和对照溪流（p<0.05），而后两者浓度差异不显著（p>0.05 ）。④对照溪流、城市上游河道和城市下游河道水质指标（如水温、pH等）之间的季节差异不同。其中，对照溪流指标季节变化较大，而城市上游河道和城市下游河道水质季节变化相对较小。 （2）城市河流底栖硅藻组成空间差异明显；相对丰度和多样性指数季节上有变化，但相对较小。①采样点的硅藻种类可以分为对照溪流、城市上游河道及城市下游河道 3个主要类群。其中，对照溪流的主要优势种为Achnanthidium minutissima；城市上游河道的主要优势种为Pseudostaurosira brevistriata和Staurosira construens var. venter；城市下游河道的主要优势种为 Nitzschia palea。主要优势种的变化特征为城市下游河道>对照溪流>城市上游河道。②主要优势种的季节变化特征表现为，A. minutissima和P. brevistriata均为春季>秋季>夏季；S. construens var. venter和 N.palea均为春季>夏季>秋季。③硅藻多样性指数种类丰富度（S）、Shannon多样性指数（H‟）、Pielou均匀度（J‟）及 Simpson多样性指数（D‟）的空间差异特征为：对照溪流>城市上游河道>城市下游河道。④不同样区的多样性指数季节变化差异较大。在对照溪流， S、H‟，J‟的季节变化特征为秋季>夏季>春季，而 D‟为秋季>春季>夏季。在城市上游河道，S的季节变化特征为秋季>春季>夏季，H‟及 D‟为夏季>秋季>春季，但 J‟为夏季>春季>秋季。在城市下游河道，硅藻多样性指数的季节变化特征均为秋季>春季>夏季。 （3）土地利用/覆盖与水质之间有较强的相关关系。①不同区域内的土地利用/覆盖类型所占比例差异较大。在河流 100 m到 2000 m的缓冲区范围内，在山区，林地是主要的土地利用/覆盖类型；在城市，建设用地是主要的土地利用/覆盖类型；而在城郊，耕地是主要的土地利用/覆盖类型。②土地利用/覆盖是北京市水质空间差异的重要影响因素。分析表明，林地与水质呈正相关关系，而不透水表面、耕地、水体等均与水质呈负相关关系。草地与水质的关系较为复杂。③土地利用/覆盖类型对河流水质的影响具有明显的尺度效应。在200 m的缓冲区范围内，林地对水质的影响大于其它土地利用/覆盖类型；而在300 m到600 m，林地、建设用地、耕地等均对水质有显著影响；而在 600 m以后，耕地及建设用地逐渐成为影响水质的主要土地利用/覆盖类型。 （4）河流底栖硅藻分布受水质影响较大，而受季节变化的影响较小。①冗余分析（RDA）表明，电导率、NH4+、HCO3-、NO3-及 Na+等5个水质因子及Chl a、平均底质组成（MSUBST）对硅藻空间分布有显著影响（p<0.05）。② RDA双序图结果表明，对照溪流、城市上游河道及城市下游河道的硅藻组成受环境影响差异明显。在对照溪流，自然底质组成、河岸较高的植被覆盖度和低浓度污染物等环境因子与A.minutissima等种类相关性较好；城市上游河道，较高比例的不透水表面和相对较低的污染物浓度等环境因子与 P.brevistriata和 S.construens var. venter等相关性较好；在城市下游河道，较高比例的不透水表面，高浓度污染物等环境因子与 N. palea等相关性较好。③典范对应分析（CCA）表明，Cl-、SiO2、NO3-、TN、WT、Cond.、Mg2+、SO42-、Ca2+、NH4+、K+及 Na+等水质指标对硅藻季节变化有显著影响（p<0.05）。④ CCA双序图结果表明,对照溪流、城市上游河道及城市下游河道的硅藻组成与环境因子之间的关系季节变化相对较小。本文有助于深入认识和理解在城市化背景下河流生态系统的变化特征和过程，并可为城市河流水环境质量的管理和调控提供重要的科学理论依据。 |
| 英文摘要 | Urbanization dramatically affects hydrology systems, water quality, and aquatic ecosystem composition in urban streams. In this study, we characterized the water quality and diatom assemblage composition along a rural to urban gradient to assess impacts of urbanization on stream ecosystems in Beijing, China. The effects of the spatiotemporal dynamics of urbanization on water quality and diatom assemblage composition were investigated using remote sensing data, field survey and statistical analysis. The analysis was conducted at two spatial scales. On the watershed scale,land use and land cover (LULC) data in four watersheds belonging to three spatial scales (rural, urban, and exurban) was obtained from TM data in 2000, 2005, and 2010. Water quality data were collected from 2000 to 2010. Relationships of the LULC types in the buffers of 100 m to 2000 m and water quality were calculated by Pearson analysis, Redundancy analysis (RDA), and multiple regression analysis. On the sampling scale, LULC data of the fifth-ring roads in the city were obtained from ALOS imagery in 2010. Water quality and diatom assemblages were collected from urban upstreams, urban downstreams, and reference streams from May to November in 2014. The spatial and temporal patterns of the water quality and diatom assemblage were quantified, and the relationships among LULC, water quality, and diatom assemblages were analyzed using multiple regression analyses. The major findings and conclusions were summarized as follows: 1. Water quality was significantly different (p<0.05) at both the watershed and sampling site level among the urban, suburban, and rural sites, but with no significant seasonal difference (p>0.05) at both scales. (1) The water quality in rural watershed was better than urban watershed and exurban watershed, the water quality of exurban watershed was the worst in the three watersheds. The concentration of BOD, for example, was 115.8 mg/L in exurban watershed, higher than 37.1 mg/L in urban watershed and 2.22 mg/L in rural watershed (p<0.001). (2) Water quality was getting better in Beijing: the concentration of pollutants was decreasing from 2000 to 2010 in urban and exurban watershed. The concentration of COD Mn was 10.7 mg/L in 2000,decreased to 8.33 mg/L in 2005 and 5.25 mg/L in 2010 (p<0.05). The concentration of pollutants in rural watershed, however, increased from 2000 to 2010 (p>0.05). (3) The water quality in urban downstream was higher than urban upstream and reference stream (p<0.05). For example, the average concentration of Cond. was 946.8 mg/L in urban downstream, higher than 463.3 mg/L in reference stream and 388.6 mg/L in urban upstream (p<0.05). But most of the water quality indicators have no significant difference between urban upstream and reference stream (p>0.05). (4) Seasonal variations of the water quality indicators have no significant difference in the sampling sites. But the seasonal variation of the water quality indicators was greater in reference stream than in urban upstreams and urban downstreams. 2. The diatom assessblages composition was very different among the sampling sites at different spatial locations (i.e., urban, suburban, and rural), but the relative abundance and the diversity index had no significant difference across season. (1) The diatom assessblages composition was divided into reference stream, urban upstream, and urban downstream by cluster analysis. Achnanthidium minutissima was the dominant species in reference stream, and Pseudostaurosira brevistriata and Staurosira construens var. venter were the dominant species in urban upstream, and Nitzschia palea was the dominant speices in reference stream. The characteristic of the dominant species relative abundance changes was greater in urban downstream than in reference stream and urban upstream. (2) Relative abundance of the species had no significant difference among spring, summer and autumn. The relative abundance‟s seasonal variation of A. minutissima and P. brevistriata was spring> autumn> summer, the seasonal variation of S. construens var. venter and N. palea was spring> summer> autumn. (3) The characteristic of the diversity index of species richness (S), Shannon diversity (H‟), Simpson (D‟) and Pielou (J‟) changes was greater in reference stream than in urban upstream and urban downstream. (4) Seasonal variation of the diversity index was different among the sampling sites. In reference stream, the seasonal variation of S, H‟, and J‟ was autumn> summer> spring,and the seasonal variation of D‟ was autumn> spring> summer. In urban upstream, the seasonal variation of S was autumn> summer> spring, the seasonal variation of H‟ and D‟ was summer> autumn> spring, but the index of J‟ was summer> spring> autumn. In urban downstream, the seasonal variation of the all index was autumn> spring> summer. 3. There was a strong correlationship between LULC and water quality. (1) The proportion of the LULC types was significant difference in the watersheds. From the buffer of 100 m to 2000 m, forest land was the main LULC type in rural watershed,impervious surface (Imp) was the main LULC type in urban watershed, and crop land was the main LULC type in exurban watershed. (2) LULC types have very different impact on water quality. Forest land was positively correlated with water quality. Pearson analysis results showed that the forest land was positively correlated with DO, but negatively correlated with NH3-N, BOD, COD Cr and COD Mn from 100 m to 2000m. However, the LULC types of water, crop, and Imp were negatively correlated with water quality. They were negatively correlated with DO, but positively correlated with NH3-N, BOD, COD Cr and COD Mn except for Imp, which was negative correlation with COD Cr from 100 m to 2000 m. (3) Multiple regression analysis results indicated that buffer sizes were very important for studying the relationship between LULC types and water quality. In buffers of 100 m and 200 m, forest land was an important factor for water quality. But in the 300 m to 600 m buffers, the mixed types of crop, impervious surface, water, and forest were significantly affected water quality. From 800 m to 2000 m, however, crop land and impervious surface were the main factors affecting water quality. 4. Water quality had great impact on the distribution of the diatom assamblages,but the seasonal variations of water quality have no significant impact on the relative abundance of diatom species. (1) Monte carlo permutation test of RDA showed that Cond, NH4+, HCO3-, NO3-, and Na+ and Chl a, MSUBST have a significant impact on the spatial patterns of the diatom assamblages (p<0.05). (2) RDA ordination diagram showed the distribution of the diatom assamblages greatly affected by the environmental indicators. In reference stream, the diatom assamblages composition correlated with high proportion of green cover near the stream, natural substrate, and low concentration of pollutants. But in urban downstream, the diatom assamblages composition correlated with high proportion of Imp cover near the stream, artificial substrate, and high concentration of pollutants. In urban upstream, the diatom assamblages composition correlated with high proportion of Imp cover near the stream, artificial substrate, and low concentration of pollutants. (3) Monte carlo permutation test of Cannonical correspondence analysis (CCA) showed that Cl-, SiO2,NO3-, TN, WT, Cond., Mg2+, SO42-, Ca2+, NH4+, K+, and Na+ have a significant impact on the seasonal variation of the diatom assamblages (p<0.05). (4) CCA ordination diagram showed the seasonal variations of the diatom relative abundance in the sampling sites have no strong correlation with the environmental change. The results were important for understanding the characteristic and process of urban stream ecosystem changes affected by urbanization. In addition, the results provide important insights on water quality regulation and control in urban streams in Beijing. |
| 内容类型 | 学位论文 |
| 源URL | [http://ir.rcees.ac.cn/handle/311016/36780]  |
| 专题 | 生态环境研究中心_城市与区域生态国家重点实验室 |
| 推荐引用方式 GB/T 7714 | 陈向. 城市化对河流生态系统的影响研究--以北京市为例[D]. 北京. 中国科学院研究生院. 2016. |
| 个性服务 |
| 查看访问统计 |
| 相关权益政策 |
| 暂无数据 |
| 收藏/分享 |
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。
修改评论